tdx-live-trading

Monitors and operates TDX live trading bridge with real-time inference, order execution, and position tracking.

1.5k|337|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/qusong0627/QuantMind --skill tdx-live-trading-qusong0627
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tdx-live-trading
Source: https://github.com/qusong0627/QuantMind/tree/main/skills/tdx-live-trading
Command: npx skills add https://github.com/qusong0627/QuantMind --skill tdx-live-trading-qusong0627

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Managing a live trading pipeline through the TDX (通达信) bridge involves many failure points: bridge disconnects, stale orders, T+1 freezes, rejected orders, and silent config issues. This Skill gives you a one-shot status snapshot plus an operational manual to diagnose anomalies and execute or cancel real orders safely. ## Core Features & Use Cases - One-Click Status Snapshot: Run a single script to check bridge health, L2 real-time inference loop, rolling strategy config, today's orders, live positions, and paper trading state. - Live Order Operations: Place, query, and cancel real orders through the TDX bridge API with correct suffix-format symbols (e.g., 600206.SH) and T+1 availability checks. - Anomaly Diagnosis: Use the built-in decision table to map snapshot symptoms (bridge offline, all-neutral scores, rejected orders, zero fees) to concrete fixes. - Use Case: During trading hours, ask the AI to run a status check every 10 minutes; when the bridge shows tdx_connected=false or orders pile up in submitted state, follow the manual to locate the fault and act. ## Quick Start Ask the AI to run the TDX live status snapshot and tell me whether the bridge, auto-trading loops, today's orders, and positions are all healthy.

Frequently Asked Questions about tdx-live-trading

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check the health of a TDX live trading bridge?

Run the tdx_live_status.py snapshot script inside the quantmind container via docker exec. It reports bridge connectivity, L2 inference loop status, rolling strategy config, today's orders, live positions, and paper trading state in one output.

How do I place or cancel real orders through the TDX bridge API?

Use the bridge HTTP API with Bearer token authentication: POST /api/v1/plans/execute to place orders, /api/v1/orders/cancel to cancel, and /api/v1/orders/query to verify final status. Stock codes must use suffix format like 600206.SH, and you should confirm available position first due to T+1 rules.

Why does my TDX order fail with codestr error?

The bridge rejects prefix-format symbols like SH600206. Use suffix format such as 600206.SH for all order, position, and query calls to the bridge.

Why is my sell order rejected with insufficient available position?

A-share T+1 rules freeze shares bought on the same day, so available_volume is 0 for today's purchases. This is expected behavior, not a bug; the position becomes sellable the next trading day.

What happens to trading when the TDX bridge goes offline?

Real order placement, position sync, and L2 real-time quotes stop when the bridge disconnects. Daily quotes, model inference, signals, and backtests continue unaffected because they use local QuantDB parquet data.

Why did my Redis config save but not take effect?

The RedisClient silently no-ops when its client is None. Standalone scripts must call r.connect() before get or set, otherwise the configuration appears saved but was never written.